Classification Of Dry Chili Based On Machine Vision

As an important ingredient,chili is increasing in demand.At the same time,as a cash crop that can effectively increase farmers' income,the planting area is also growing.The classification of chili appearance quality is still relatively outdated.In order to improve the economic benefits of dry chili and reduce the cost of dry chili classification,the classification system was designed based on machine vision.The system that took DSP6000 microprocessor as processor,included lighting system,digital camera,and feeding part.The dry chili classification system consisting of the material device,the conveying device and the pneumatic sorting device performs real-time collection and processing on the appearance image of the dry chili,and completes the real-time classification treatment of the dry chili according to the appearance quality characteristics of the dry chili extracted by the system.(1)Obtain the appearance parameters of dry chili by measurement,determine the physical properties such as length,width and quality of dry chili,and determine the specific classification standard of dry chili according to relevant national standards.(2)Build a machine vision system,design the image acquisition background,illumination system and display mode of the machine vision system for the appearance quality image acquisition process of dry chili,and process the appearance quality image of dry chili,using line section method,minimum an algorithm such as circumscribed rectangle extracts the color and shape characteristics of dry chili,a method based on the watershed algorithm to segment multiple dry chili images and use a naive Bayesian algorithm to simultaneously classify multiple dry chili is proposed.The classification accuracy rate is over 95%,and the classification effect can reach the experimental expectation.(3)According to the actual needs and experimental needs,design and trial production of dry chili classification selection experimental device,the design principle and specific parameters of the three main components of the experimental device were expounded.Through the whole machine experiment,the feasibility of the algorithm in the study and the operational reliability of the dry chili classification device were verified.(4)After calculation and experimental verification,the optimal tilt angle of the dry chili feeding device was obtained;the running time of the algorithm was optimized,and the speed of the conveyor belt in the classification experimental device was optimized to optimize the classification efficiency of the dry chili.